Accurate diagnosis of sepsis using a neural network: Pilot study using routine clinical variables

Arriaga-Pizano, Lourdes; González-Olvera, Marcos A.; Ferat‐Osorio, Eduardo; Escobar, Jésica; Hernández-Pérez, Ana Luisa; Revilla-Monsalve, Cristina; Lopez-Macias, Constatino; León-Pedroza, José Israel; Cabrera-Rivera, Graciela Libier; Guadarrama-Aranda, Uriel; Leder, Ron S.; Gallardo-Hernández, Ana Gabriela

doi:10.1016/j.cmpb.2021.106366

Cited by 4 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… 12 It is characterized by persistent maladjusted inflammation and immunosuppression resulting from disturbed immune homeostasis caused by invading pathogens. 13 The early signs of sepsis are not specific; 14 therefore, it is crucial to identify novel biomarkers for the early classification of sepsis to permit timely treatment to reduce the sepsis mortality rate. Our goal is to help classify septic patients earlier and manage them more promptly when they are septic or at risk of septic shock.…”

Section: Discussionmentioning

confidence: 99%

The Role of AGGF1 in the Classification and Evaluating Prognosis of Adult Septic Patients: An Observational Study

Ji,

Wan,

Zhang

et al. 2024

IDR

View full text Add to dashboard Cite

Purpose Angiogenic factor with G patch and FHA domains 1 (AGGF1) is a crucial angiogenic factor that is involved in a variety of diseases and in the regulation of inflammatory responses. However, its role in sepsis is poorly understood. We have investigated the role of AGGF1 in the classification and prognostic evaluation of adult septic patients in a clinical context. Patients and Methods A total of 126 septic patients who visited the Emergency Department of Beijing Chao-Yang Hospital and 76 non-sepsis patients visiting the Physical Examination Center of Beijing Chao-Yang Hospital were enrolled. AGGF1 levels in plasma were detected by enzyme-linked immunosorbent assay. Spearman correlation analysis was used to determine correlations between plasma AGGF1 and Sequential Organ Failure Assessment (SOFA) score, Acute Pathology and Chronic Health Evaluation II (APACHE II) score, procalcitonin and lactate. We evaluated the classification significance of AGGF1 in sepsis using receiver operating characteristic (ROC) curves. We also assessed the predictive significance of AGGF1 for 28-day mortality in sepsis using ROC curves and Kaplan–Meier analyses. Results Plasma AGGF1 levels were higher in sepsis patients than in non-sepsis patients (P < 0.001). Among sepsis patients, plasma AGGF1 levels were higher in non-survivors than in survivors (P < 0.001). Increased plasma AGGF1 levels were positively correlated with SOFA score, APACHE II score, procalcitonin and lactate. Plasma AGGF1 levels could distinguish sepsis patients from non-sepsis patients (area under the curve [AUC] = 0.777). AGGF1 had a higher predictive value than SOFA score, APACHE II score, lactate, procalcitonin, and white blood cell count for 28-day mortality in patients with sepsis (AUC = 0.876). Furthermore, Kaplan–Meier analysis indicated that lower plasma AGGF1 levels were associated with lower 28-day mortality compared with higher plasma AGGF1 levels (log rank P < 0.001). Conclusion AGGF1 is useful for the classification and evaluating prognosis of adult septic patients.

show abstract

Section: Discussionmentioning

confidence: 99%